JP2006023295A - Birefringence-measuring method and birefringence measuring apparatus using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To carry out a precise and quantitative birefringence measurement, and to measure a wide measurement plane at one time, in order to make the measurement speedy. <P>SOLUTION: In a birefringence-measuring method, a light beam from a light source is made uniform substantially and circularly-polarized, and is brought to impinge on the object to be tested, and then a plurality of images are acquired by a CCD camera as a first image set, wherein the variations in brightness produced by rotating a linear polarization plate relatively to a light beam passing through the object to be tested are recorded by pixels; a plurality of images are acquired by the CCD camera as a second image set, wherein the variations in brightness produced by rotating the linear polarization plate at the same rotation angle without inserting the object to be tested are recorded by pixels; and then the first image set and the second image set are compared with each other on a pixel to pixel basis, thereby obtaining birefringence values and optical axes of the object to be tested at a plurality positions in the measurement plane. In the birefringence-measuring method, at least the first and second image sets are subjected to a process, wherein the brightness signals of the CCD camera are converted into light intensity values, and a compensation process is carried out, based on the transmissivity of the object to be tested and the polarization degree and the extinction ratio of the light beam impinging on the object to be tested. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a birefringence measuring method and apparatus for an optical test object made of a material such as glass or plastic.

  With the recent increase in the density and accuracy of optical elements due to the development of optical technology, there is a possibility that the performance degradation of the optical elements due to birefringence becomes a problem.

  For example, molded lenses made of plastic materials are widespread. Plastic molding has the advantage that even complex shaped parts can be produced with high efficiency, but has the disadvantage that birefringence occurs inside the machined part due to residual strain generated during the machining process.

  Further, in an imaging optical system mounted on a digital camera or a mobile phone, downsizing and high definition are required, and a zoom lens using an optical element as disclosed in Patent Document 1 is also expected. The optical element of Patent Document 1 forms an image by repeatedly reflecting light inside an element formed of a transparent body such as plastic or glass. Also in this case, if the birefringence inside the element is large, it may cause a deterioration in imaging performance.

  Therefore, it is necessary to accurately grasp the influence of the optical element on the polarization, and to suppress the deterioration of performance due to birefringence within an allowable range. For this purpose, it is necessary to accurately measure the polarization characteristics of each optical element. Furthermore, in view of mass productivity and development efficiency, it is desirable that these birefringences be measured at high speed.

  On the other hand, with respect to the birefringence measurement method, polarization measurement results in measuring the ellipticity and principal axis direction, or the amplitude ratio and relative phase difference in two orthogonal coordinate systems. Well-known measurement methods can be classified into a method of measuring by modulating the phase and a method of measuring by rotating the analyzer. In the phase modulation method, the irradiation light is modulated using a photoelastic modulator, and birefringence is obtained from the phase of the beat signal and the modulation signal of the light transmitted through the transparent test object. In the rotating analyzer method, as described in Patent Document 2 and Patent Document 3, transmitted light is received by the light receiving element on the back surface of the analyzer while rotating the analyzer placed on the back surface of the transparent object, The birefringence is to be obtained by the change in the light receiving output from the light receiving element accompanying the rotation. The apparatus described in Patent Document 3 is a video camera in which a test object is arranged between a polarizer and an analyzer arranged in a crossed Nicols mode, and the amount of light transmitted through these is rotated by the crossed Nicols. It is a configuration to capture. In the measurement, the maximum value of the detected light intensity and the rotation angle when the maximum value is output are obtained, and the amount of birefringence of the test object is obtained as a function of the rotation angle of the crossed Nicols. In the measurement method described in Patent Document 3, the polarization of the light beam incident on the test object is changed to circularly polarized light by the polarizer and the ¼λ plate, and similarly, from the change of the rotation angle of the analyzer and the detected light intensity. The amount of refraction is obtained.

  In these measurement methods, a transparent parallel object is irradiated with a thin parallel beam, light transmitted from the object is received by a light receiving element such as a photodiode, and the polarization state of the transmitted light due to birefringence of the object is measured. The birefringence of the test object is obtained by detecting the change.

  Further, in Patent Document 4 and Patent Document 5, the test object is irradiated with expanded parallel light, and the transmitted light is received by a two-dimensional sensor such as a CCD sensor, thereby obtaining the birefringence of the test object. The birefringence surface measurement is possible.

In Patent Document 6, a configuration in which a phaser can be moved back and forth between a test object and a rotating analyzer is used, and two measurement results in which only the analyzer is rotated with and without the phaser being compared are compared. As a result, the measurement range is expanded from −λ / 4 to λ / 4 to −λ / 2 to λ / 2 from the analysis in the long axis direction of the elliptically polarized light of the light reaching the CCD.
Japanese Patent No. 3292173 JP-A 63-269045 JP-A-6-147986 JP-A-4-58138 JP-A-7-77490 Japanese Patent No. 3456695

  However, in Patent Document 2 and Patent Document 3, the light intensity with respect to the rotation of the orthogonal Nicols or the analyzer has to be measured at a considerably large number of points, and there is a problem that the measurement takes time. Furthermore, in Patent Document 2, since the specimen is irradiated with linearly polarized light, if the polarization direction coincides with the direction of the axis of the specimen, the polarization state is transmitted without being changed. Birefringence information cannot be detected.

  In Patent Documents 4 and 5, the measurement time is shortened by surface measurement. However, a large-diameter polarizer and a wave plate are required to measure an object having a wide aperture at once. . However, in reality, the uniformity of the large-diameter wave plate is poor, and it is difficult to align the measurement surface in a uniform polarization state, and this causes a decrease in the measurement accuracy of birefringence. In addition, it is difficult to illuminate the measurement surface with uniform brightness even with illumination light, and this non-uniformity causes a decrease in measurement accuracy in detecting the light intensity with respect to the rotation of the analyzer.

  Further, in Patent Document 6, a phase difference of λ / 4 is given to transmitted light through a λ / 4 plate which is a phase shifter with respect to the light transmitted through the test object, and the elliptical polarized light in the major axis direction at this time is given. From the change, the rotation direction of the polarization of the light transmitted through the test object is substantially obtained. For example, the rotation direction of the polarization is different between the phase difference of 0 to λ / 4 and λ / 4 to λ / 2. The range of relative phase difference that can be measured by using it is expanded. However, in an actual test object, the direction of the optical axis is generally different depending on the part of the element. Originally, in order to give a phase difference of λ / 4 to the light transmitted through the test object, a desired angle such as 45 degrees of the optical axis of the λ / 4 plate with respect to the direction of the optical axis of the measurement point of the test object. Must be placed in. Therefore, when the method of Patent Document 6 is used, a phase difference of λ / 4 cannot always be given to an object whose optical axis direction differs depending on the portion of the element. It becomes difficult to judge accurately.

  On the other hand, regarding the accuracy of the measured value, as described in Patent Document 3, the amount of birefringence, that is, the relative phase difference δ is determined by the principal axis direction φ at the measurement point, the rotation angle θ of the analyzer, the incident light intensity Io, and the transmitted light intensity I. A trigonometric function is expressed as equation (3) described below. In actual measurement, the intensity ratio I / Io is measured while changing the rotation angle θ of the analyzer. The change in the intensity ratio I / Io at this time is shown in FIG. In FIG. 5, the solid line is the theoretical curve of formula (3), and the round points are measured values. Equation (7) described later is obtained by solving this equation (3) with respect to the relative phase difference δ, and the graph is shown in FIG. As can be seen from FIG. 12, the sensitivity of the relative phase difference δ to the minute change in I / Io differs depending on the value of the intensity ratio I / Io. Since the measurement intensities I and Io are observed as discrete values according to the number of bits of a light receiver such as a CCD, the measurement accuracy is several in the vicinity of δ = 0 in the vicinity of δ = ± λ / 4 in FIG. There was a problem that it deteriorated from several times to several hundred times.

  Therefore, in order to measure the birefringence of the test object in a wide range and with high accuracy, the measurement surface is measured in consideration of the illumination unevenness and the degree of polarization, and the main axis direction (optical axis) at each measurement point is different. The phase must be shifted corresponding to the direction of

  In view of the above points, the present invention uses a birefringence measurement method capable of measuring a wide measurement surface at a time while maintaining a high measurement accuracy and a wide measurement range with respect to the amount of birefringence and the principal axis direction of the test object and the same. The aim is to propose a device.

  In the birefringence measurement method of the present invention, a light beam from a light source is made substantially uniform by a diffuser plate and is circularly polarized, and is irradiated on a test object, and a linearly polarizing plate is applied to the light beam transmitted through the test object. A plurality of images recorded for each pixel by a CCD camera that show changes in brightness appearing by rotating are used as the first image, and appear by rotating the linearly polarizing plate with the same rotation width without inserting the test object. A plurality of images in which changes in brightness are recorded for each pixel by a CCD camera are used as a second image, and the first image and the second image are compared for each pixel, so that the amount of birefringence and optical properties of the test object are compared. In a birefringence measurement method for obtaining an axis at a plurality of locations in a measurement surface, at least first and second images are converted from luminance signals of a CCD camera into light intensity, and the transmittance of the test object and the test object are irradiated. Based on the degree of polarization and extinction ratio And performing a positive process.

  The birefringence measuring method of the present invention also converts circularly polarized light (for example, non-polarized light emitted from a light source into linearly polarized light through a polarizing plate, and converts the linearly polarized light into circularly polarized light through a λ / 4 plate or the like. The first and second changes are recorded on a pixel-by-pixel basis using a CCD camera to record the change in light and darkness that appears when the linearly polarizing plate (analyzer) is rotated with respect to the light beam that has passed through the object. A plurality of images obtained by measurement are set as a first image, and a birefringence amount of the test object is determined from a waveform of a change in received light intensity with respect to a rotation angle of a linear polarizing plate (analyzer) in a measurement region having each pixel as a minimum unit. In the birefringence measurement method for determining the direction of the optical axis at a plurality of locations in the measurement plane, a wave plate having a known phase difference is inserted between the test object and the linear polarizing plate (analyzer), and the optical axis of the wave plate And keep the transmission axis of the linear polarizer (analyzer) at a predetermined angle. A plurality of images obtained by the second measurement in which the change in brightness appearing by rotating at the same time is recorded for each pixel by the CCD camera is defined as a second image, and the first image and the second image are defined for each pixel. In comparison, the amount of birefringence of the test object and the direction of the optical axis are measured.

  Also, the waveform of the intensity change of the transmitted light obtained in the first measurement is compared with the waveform obtained in the second measurement, and the measured value of the birefringence obtained in the first measurement is the test object. Determine whether or not the range of measurable birefringence exceeds the range of measurable birefringence due to periodic changes to the actual amount of birefringence. In the case where the measured birefringence amount is added to the measured birefringence amount in accordance with the periodic change, the measurement range is expanded, and the birefringence value obtained by the first measurement is in a region where the measurement accuracy is low. In the case of the birefringence amount, a value obtained by subtracting the phase difference of the inserted wave plate from the second measurement result is used as the birefringence amount at the measurement point.

  The wave plate inserted in the second measurement is a λ / 4 plate.

  In addition, the transmission axis of the linearly polarizing plate (analyzer) and the fast axis of the wave plate inserted in the second measurement are inclined by 45 degrees.

  Furthermore, the birefringence measuring method of the present invention is a linear polarizing plate in which a light beam from a light source is made uniform by a diffusion plate, the light beam is made circularly polarized by a phase difference plate (λ / 4 plate), and a test object is not inserted. A plurality of images obtained by the first measurement in which a change in brightness appearing by rotating the (analyzer) is recorded for each pixel by a CCD camera is defined as a first image, and a circularly polarizing plate and a linearly polarizing plate (analyzer) The test object is inserted between the two), and the change in brightness and darkness that appears when the linearly polarizing plate (analyzer) is rotated at the same rotation width for the light beam transmitted through the test object is recorded for each pixel by the CCD camera. By comparing the first image and the second image for each pixel, the birefringence amount of the test object and the direction of the optical axis can be determined in the measurement plane. In the birefringence measurement method obtained at a plurality of locations, the first and second images are connected. The brightness signal of the CCD camera is converted into light intensity, and correction processing is performed based on the transmittance of the test object and the polarization degree and extinction ratio of the light beam that irradiates the test object. A wave plate with a known phase difference is inserted between the photons), and light and dark appear by rotating simultaneously while maintaining the optical axis of the wave plate and the transmission axis of the linearly polarizing plate (analyzer) at a predetermined angle. A plurality of images obtained by a third measurement in which changes are recorded for each pixel by a CCD camera are used as a third image, and the first image, the second image, and the third image are compared for each pixel. Thus, the birefringence amount of the test object and the direction of the optical axis are measured.

  Further, the waveform of the intensity change of the transmitted light obtained in the second measurement is compared with the waveform obtained in the third measurement, and the measured value of the birefringence obtained in the second measurement is the test object. Determine whether or not the range of measurable birefringence exceeds the range of measurable birefringence due to periodic changes to the actual amount of birefringence. The measurement range is expanded by taking into account the amount corresponding to the periodic change in the amount of birefringence to be measured, and the value of the birefringence obtained by the second measurement is in a region where the measurement accuracy is low. When the amount is birefringence, the value obtained by subtracting the phase difference of the inserted wave plate from the third measurement result is used as the amount of birefringence at the measurement point.

  The dark level based on the extinction ratio of the measurement optical system and the dark level of the CCD camera are incorporated in a conversion table or conversion function for converting the luminance signal of the CCD camera into light intensity. .

  The first and second images are converted from the luminance signal of the CCD camera into light intensity, and correction processing is performed based on the transmittance of the test object, the degree of polarization of the light beam irradiating the test object, and the extinction ratio. The light and dark appearing by inserting a wave plate having a known phase difference between the test object and the linear polarizing plate, and rotating simultaneously while maintaining the optical axis of the wave plate and the transmission axis of the linear polarizing plate at a predetermined angle. When a plurality of images obtained by the third measurement in which changes are recorded for each pixel by a CCD camera are used as the third image, the first image and the second image, or the first image and the third image By comparing the images for each corresponding pixel, it is characterized in that it is not affected by uneven illumination and noise depending on the measurement position by a CCD or the like.

  The light source is characterized by using a primary color filter for white or white.

  In addition, the light source in the previous period has a bright line, and the CCD camera can receive light in color.

  Further, the measurement by the measurement method is controlled from the PC monitor screen.

  The correction of the transmittance and the degree of polarization of the test object is performed by changing the light intensity ratio between the first image and the second image or the first image and the third image with respect to the rotation of the analyzer at the measurement point. The correction is performed based on an amount of deviation of the average value of the maximum value and the minimum value from a desired value.

  Further, for correcting the transmittance and the degree of polarization of the test object, a ratio I / Io of the light intensity I of the first image to the light intensity I of the second image Io with respect to the rotation of the analyzer at the measurement point. When the average value of the maximum value and the minimum value of the change is Iave, the birefringence amount and the direction of the optical axis are obtained from the light intensity ratio (I / Io) ′ corrected by the following equation.

T = Iave / 0.5
(I / Io) ′ = [{I / Io + (1−T) / 2} −0.5] /T+0.5
Further, in addition to the measurement optical path, it can be converted into a fiber having a diffusion plate and / or a polarizing plate at the tip branched from the measurement optical path, and the optical axis passing through the exit surface of the fiber is the optical axis of the CCD camera. It is characterized by not matching.

  Furthermore, the birefringence measuring apparatus of the present invention is characterized by using the above-described birefringence measuring method.

  Furthermore, the optical element of the present invention is characterized in that the amount of birefringence and the direction of the optical axis are measured by using the above-described birefringence measuring apparatus, and the optical performance is compensated by a processing condition or a correction element.

  Furthermore, the optical system of the present invention is characterized by using the above-described optical element.

  Here, the light intensity refers to the intensity of light incident on the CCD itself, and a gradation signal corresponding to the number of bits of the CCD is used as a luminance signal.

  The following four factors are particularly cited as factors that lower the measurement accuracy when performing birefringence surface measurement of a test object. The first factor is the sensitivity characteristic of the light receiving element, the second factor is the illumination unevenness of the measurement surface, the third factor is the transmittance of the test object, and the fourth factor is the manufacture of each element of the measurement optical system. It is an error.

  As for the first factor, when a light receiving element such as a CCD is used, the camera itself often has a γ characteristic. As shown in FIG. 9, the output luminance signal and the input light intensity according to the number of bits of the camera. Is a nonlinear intersystem. Therefore, it is necessary to correctly convert the output luminance signal of the camera into the input light intensity. Here, a mark indicated by a square point in FIG. 9 is a measured value, and it is approximated that a linear change occurs between the measured points. Since approximation is included in the interpolation of measurement points, increasing the number of measurement points in such a conversion table affects the improvement of measurement accuracy. Further, the black level of the camera is determined when creating this conversion table.

  Regarding the second factor, in the measurement principle of the present invention, the change in the light intensity ratio I / Io between the blank image and the test object image measured with the test object inserted at a certain measurement point with respect to the rotation of the analyzer is determined. Therefore, when the same value of Io twice the light intensity of the blank image is used at all measurement points, partial unevenness of illumination and variations due to the position of the light intensity due to the image plane illuminance distribution affect the measurement accuracy. In order to correct such illumination unevenness, the light intensity ratio I / Io may be calculated for each range of measurement points with each pixel as the minimum unit. As a result, it is possible to perform measurement that is not easily affected by unevenness due to other factors, including an illuminance distribution that can be easily predicted and expressed as a function.

  Regarding the third factor, the optical element of the test object is often a transparent body, but the transmittance of the test object is not 100%. This decreases the value of the light intensity I of the test object image, so that the value of the light intensity ratio I / Io changes, and the measurement accuracy deteriorates as the transmittance decreases. There is a method of measuring the transmittance of the test object in advance and correcting the value of I, but considering the case where the thickness of the test object is partially different and the transmittance is also partially different, it is not preferable in terms of measurement efficiency. . A correction method in consideration of the transmittance of the test object without reducing the measurement efficiency will be described together with the correction of the fourth factor described below.

  Regarding the fourth factor, the measuring device of the present invention is characterized in that a large area can be measured in a lump, but it requires a large polarizing plate and a wave plate as well as a wide measuring surface. In general, when a large wavelength plate having a diameter exceeding 150 mm is used, it is difficult to uniformly generate a desired phase difference over the entire surface. For this reason, even if it is going to produce | generate circularly polarized light with a polarizing plate and (lambda) / 4 plate, it does not become perfect circularly polarized light. As a problem due to this, there is a contradiction in performing measurement that is not affected by the principal axis direction of the test object by irradiating the test object with circularly polarized light. Further, since the extinction ratio is deteriorated, the black level of the measurement system is slightly increased and the measurement accuracy is deteriorated. Regarding the extinction ratio, if a λ / 4 plate made of crystal or the like is arranged with high accuracy is placed on the object table, the light transmitted through the λ / 4 plate should be linearly polarized. The black level is determined from the image in the extinction state with respect to the rotation, and may be taken into consideration when creating the light intensity conversion table of the camera in advance. On the other hand, regarding the irradiation light to the test object becoming elliptically polarized light, a relative phase difference appears large in a certain direction determined by the direction of the ellipse, and appears small in a direction orthogonal thereto. This is because the double Io of the light intensity of the blank image is not constant with respect to the rotation of the analyzer, and the change in the light intensity ratio I / Io cannot be correctly obtained due to the influence of the principal axis direction. A method for correcting the transmittance of the test object at the same time will be described below. When the transmittance of the test object is not 100%, the change in the light intensity ratio I / Io with respect to the rotation of the analyzer obtained in this measurement is as shown in FIG. By transmitting the test object with respect to twice the light intensity Io at each measurement point obtained from the blank image, the image of the test object becomes darker by the transmittance over the entire rotation of the analyzer. Therefore, as shown in FIG. 10, the vibration center Iave of the waveform of the light intensity ratio I / Io falls below I / Io = 1/2. In this state, for example, if the relative phase difference is calculated from the value of Imax or the value of Imin, an incorrect result is obtained. The same applies to the main axis direction. Similarly, when the irradiation light to the test object is slightly elliptically polarized, the center of the amplitude is similarly deviated from I / Io = 1/2. Therefore, the correction amount may be determined from the amount of deviation from I / Io = 1/2 of the vibration center Iave of the obtained waveform. Thereafter, the value shown in the equation (1) determined from this amount of deviation is the transmittance error correction coefficient. Call it T.

T = Iave / 0.5 (1)
The correction method is as follows: the light intensity ratio I / Io is (0.5-Iave) and rewritten with the transmittance error correction coefficient T, (1-T) / 2 is added, and the vibration center Iave of the waveform is I / Io = 1. Shift to / 2. Further, when the transmission intensity is lowered, the amplitude of vibration Iamp in FIG. That is, the light intensity ratio at each measurement point may be corrected according to the equation (2).

(I / Io) ′ = [{I / Io + (1−T) / 2} −0.5] /T+0.5 (2)
FIG. 11 shows the corrected waveform. Various amounts after correction using the transmittance error correction coefficient T are indicated with a prime.

  As described above, the image data is used to correct factors caused by measurement errors such as CCD sensitivity characteristics, illumination unevenness, degree of polarization, and transmittance of the test object. Even with a large area, measurement with high measurement accuracy is possible.

  Here, in addition to the above four measurement accuracy degradation factors, there is a degradation in measurement accuracy that contributes to the Sin function shown in FIG. A means for solving this will be described. A wave plate such as a λ / 4 plate having a known phase difference is inserted between the test object and the analyzer, and the optical axis of the wave plate and the transmission axis of the analyzer are kept at an angle of +45 degrees. The change of brightness and darkness that appears by rotating simultaneously is recorded for each pixel by the CCD camera. Since the principal axis direction of the test object has already been obtained from the comparison of the image transmitted through the test object without the wave plate inserted and the image excluding the test object, the wave plate is inserted. At a certain measurement point, there is always one image in the state where the fast axis of the λ / 4 plate is orthogonal to the principal axis direction of this measurement point in the group of images captured for each rotation angle pitch of the analyzer. To do. Since this is the same for each pixel, the angle between the analyzer and the axis of the λ / 4 plate is +45 degrees, and the phase is shifted by + λ / 4 by rotating and measuring at the same time. It is possible to obtain the value by a single measurement over the entire surface of the test object having different main axis directions for each part.

  As described above, since the phase can be shifted by + λ / 4 over the entire surface of the test object, the polarization of the light transmitted through the measurement point of the test object from the change in the measured value before and after giving the phase difference. Thus, the birefringence amount that is periodically measured and measured with respect to the actual birefringence amount can be doubled in consideration of the polarization rotation direction. Furthermore, in the birefringence measurement method using the circularly polarized light as the irradiation light, the measurement accuracy is most deteriorated when the birefringence amount of the test object is an integral multiple of λ / 4, but this also shifts the phase by + λ / 4. By doing so, it becomes possible to perform measurement with the highest measurement accuracy. This is because there is only one point where the phase of the measurement point can be shifted by λ / 4 while the analyzer rotates with the λ / 4 plate inserted, but this one point is necessary for calculating the birefringence amount. This is because it is also a point.

  As described above, according to the present invention, it is possible to quantitatively measure a wide measurement surface at a time while maintaining high measurement accuracy for the amount of birefringence and the principal axis direction of the test object.

  Before showing measurement examples of the present invention, the measurement principle of the measurement method of the present invention and the configuration of a measurement apparatus using the same will be described.

  The measurement principle of the embodiment of the present invention will be described with reference to FIG. 2, and the configuration of the birefringence measurement apparatus will be described with reference to FIG. In FIG. 2, the measurement method is as follows. In the optical path between a white light source and a two-dimensional light receiving surface such as a CCD, a test object 1 is a polarizer 4 and a λ / 4 plate disposed with an axis inclined by 45 with respect to the polarizer 4. A circular polarizer composed of a (retardation plate, wave plate) 5 (an optical system that emits circularly polarized light, a light source that emits linearly polarized light, and an optical axis inclined by 45 degrees with respect to the polarization direction) / 4 plate) and an analyzer (linear polarizer, an optical element that transmits only predetermined linearly polarized light and shields linearly polarized light orthogonal thereto) 7. A plurality of gray images on the CCD surface with respect to the rotation of the analyzer 7 are sampled. When the test object has no birefringence, the analyzer 7 samples with respect to the circularly polarized light, so that ideally, the luminance signal does not change with the rotation of the analyzer. When the test object has birefringence, since the light transmitted through the test object is elliptically polarized, a luminance signal change with respect to the rotation of the analyzer occurs.

  FIG. 1 shows a schematic diagram of a measuring apparatus equipped with this optical system. In FIG. 1, a light source 2 is a white light source such as a halogen lamp, and is converted into parallel light by a lens 11, and a folding mirror 12 is inserted into and removed from an optical path, whereby a light beam is directed toward a cover glass 6 that is a measurement stage. The direction can be switched. The light flux returned by the return mirror 12 is enlarged by the condensing lens 11 and irradiated on the diffusion plate 3. The light beam diffused by the diffusing plate 3 becomes circularly polarized light by the linearly polarizing plate 4 and the λ / 4 plate 5 and enters the test object 1 through the cover glass 6 as a measurement stage. The light beam emitted after the polarization state is changed by the test object 1 having birefringence passes through the analyzer 7 and is received by the CCD camera 9 having the imaging lens 8. At this time, the focal position of the CCD camera 9 is in the vicinity of the object surface to be examined, the analyzer 7 and the CCD camera 9 can be controlled by the PC 13, and the rotation angle of the analyzer 7 is controlled by a stepping motor or the like. Is stored in the PC 13. The polarizing plate 4 and the wave plate 5 have an angle memory on the holding frame, and the angle can be easily adjusted. Further, in order to improve the measurement accuracy and expand the range described later, a detachable wave plate 7 ′ is disposed between the test object 1 and the analyzer 7, and is disposed adjacent to the test object side of the analyzer 7, The analyzer 7 and the wave plate 7 ′ can be rotated together by the motor.

  On the other hand, when the folding mirror 12 is removed from the optical path, the light beam from the light source 2 is put into the fiber 10 by the condenser lens 11. Similarly, circularly polarized light is generated by the diffusion plate 3, the linear polarizer 4 and the λ / 4 plate 5 disposed in the vicinity of the exit end of the fiber. By using such a fiber, it is possible to illuminate an optical element having a different incident direction and emitting direction as shown in FIG. 3 from an arbitrary direction. The optical element of FIG. 3 is made of a transparent material such as glass or plastic, and light incident from the incident surface is emitted from the exit surface after being repeatedly reflected a plurality of times. Since the light beam passes through the inside of the transparent body from the incident surface to the emission surface, light having a relative phase difference is emitted.

  Further, the wavelength of the measurement light can be measured at a wavelength that matches the measurement purpose by inserting a wavelength selection element such as a color filter 15 or a band pass filter into the optical path in the illumination system or the imaging system.

  However, when such a measuring device is actually assembled, a highly accurate circular polarizer is required to enable measurement of a large-diameter test object having a diameter exceeding 150 mm in a large area at once. Is very difficult, the light transmitted through the circular polarizer becomes elliptically polarized light deviated from perfect circularly polarized light. Even when a small circular polarizer with high accuracy is used at a narrow position of the light beam, since the light beam is broadened with a lens in order to measure in a large area, the uniformity of the polarization state deteriorates. For this reason, the polarization state of the light incident on the test object deviates from the desired state, which is a major factor that degrades the measurement accuracy. For this reason, conventionally, a method of repeating measurement in a very small area using a laser beam has been performed, and in the case of a large area, high-precision measurement could not be performed.

  In addition to the disturbance of the polarization state due to such optical elements, illumination light intensity unevenness on the measurement surface by the illumination optical system and illumination unevenness such as the distribution of the illuminance ratio of the image plane are also major factors that degrade the measurement accuracy. Become. Further, the obtained grayscale image has the characteristics of the output luminance signal with respect to the incident light intensity depending on the number of bits of a light receiver such as a CCD. Therefore, the luminance signal change of these images is converted into the light intensity as accurately as possible. It is also important.

  Accordingly, the PC 13 calculates the birefringence amount and the principal axis direction of the test object by performing the correction processing of the present invention for the factors that degrade the measurement accuracy from a plurality of accumulated images, and displays them on the monitor 14. Information displayed on the monitor includes the amount of birefringence at each position on the measurement surface with each pixel as the minimum unit and a text data string in the principal axis direction, the amount of birefringence at each point on the measurement surface, and the principal axis direction. 2D distribution chart for easy visual understanding by color and luminance signal or arrow according to direction, 3D distribution chart showing birefringence amount in height direction, luminance signal for analyzer rotation at each measurement point Or it is a graph etc. which show the change of light intensity.

  Further, the calculation and display of the birefringence amount and the principal axis direction can be output as an average or mode value in an arbitrary range although the range corresponding to each pixel is a minimum unit.

  Next, details of the measurement principle and image processing of the grayscale image stored in the PC will be described. The intensity I of the light transmitted through the circular polarizer, the test object, and the analyzer is given by Expression (3).

I / Io = 1/2 · {1 ± sin2 (θ−φ) sinδ} (3)
Here, Io is the intensity of light incident on the test object, θ is the rotation angle of the analyzer, φ is the principal axis direction, and δ is the relative phase difference. In Expression (3), ± is + when using clockwise circularly polarized light, and − when using counterclockwise circularly polarized light. The main axis direction φ is defined as a range of 0 ≦ φ ≦ π.

  As can be seen from Equation (3), sin δ is a constant that depends on the relative phase difference δ for each measurement point of the sample and contributes to the amplitude with respect to the rotation of the analyzer at a certain measurement point. Further, sin2 (φ−θ) is a term that changes when −1 ≦ sin2 (φ−θ) ≦ 1 and contributes to the cycle. Therefore, the center of the amplitude in the equation (3) is 1/2, the maximum amplitude is sin2 (φ−θ) = 1 (sinδ / 2), and the period is π.

  FIG. 4 shows an example of an image accumulated in the PC when the analyzer is rotated to π (180 degrees) with θ = 0 as the direction of the transmission axis at the start of measurement. At this time, the reference direction for the rotation angle θ = 0 of the analyzer may be determined so that the measurer can easily understand the apparatus. The image in FIG. 4 is a schematic diagram assuming a test object having a relative phase difference δ of + π / 4 at all measurement points and a principal axis direction φ of π / 8.

  The blank image in FIG. 4 is an image sampled without placing a test object in the measurement optical system. The test object image is an image obtained by arranging the test object and sampling it. The number of sampling points is related to the fitting accuracy in equation (3). When attention is paid to a certain point on the test object image, the change in the ratio I / Io between twice the light intensity Io of the blank image and the light intensity I of the test object image at each θ is as shown in FIG. The horizontal axis in FIG. 5 is the rotation angle θ of the analyzer, the vertical axis is the light intensity ratio I / Io, the black circle is the measured value, and the solid line is the fitting curve of Equation (3).

  Next, the derivation of the amount of birefringence will be described. Deriving the amount of birefringence is nothing but finding the relative phase difference δ at the measurement point of the test object. Further, the relative phase difference δ can be easily obtained from the amplitude of the fitting curve.

  By fitting the change in the light intensity ratio I / Io with respect to the angle of rotation θ of the analyzer at each measurement point using the function of Equation (3), the birefringence amount is obtained from the amplitude and the principal axis direction is obtained from the phase by the following calculation. It is done.

  This curve takes the maximum value (I / Io) max when sin δ> 0, and when sin 2 (θ−φ) = 1, that is, (θ−φ) = π / 4, and sin δ <0. When sin2 (θ−φ) = − 1, that is, (θ−φ) = 3π / 4, Equation (4) is obtained.

(I / Io) max = 1/2 · {1 + sin δ} (4)
Therefore, when paying attention to one measurement point, the sign of sin δ and the sign of sin 2 (θ−φ) cannot be obtained simultaneously from the fitting curve of I / Io. Therefore, this can be solved by, for example, determining an arbitrary one point in the entire measurement range such as the center of the measurement surface as a reference point for the amount of birefringence and determining the relative phase difference δ at this point as, for example, positive. Since δ is determined to be positive, the sign of sin 2 (θ−φ) at this point is determined, so that the main axis direction φ can be determined. Thus, it is difficult to assume that the reference point is provided so that the relative phase difference and the main axis direction between adjacent pixels (measurement points) change so rapidly that the sign of sin δ or sin 2 (θ−φ) is reversed. Considering this, the relative phase difference can be measured in the range of −π / 2 ≦ δ ≦ π / 2 from the continuity of the measurement values between adjacent pixels.

  Thus, the relative phase difference δ is obtained from the maximum value of the fitting curve. Further, the relationship between the relative phase difference δ and the amount of birefringence can be expressed by Equation (5) using the measurement wavelength λ.

Birefringence amount = (δλ) / (2π) (5)
In the above example shown in FIG. 5, since the amplitude when sin2 (θ−φ) = 1 is 0.85355, it can be seen from equation (4) that δ = + π / 4.

  Next, derivation in the main axis direction will be described. The main axis direction here is defined as the direction of the vibration surface of the fast wave.

  When the rotation angle θ coincides with the main axis direction φ of the test object while rotating the analyzer by 180 degrees, (θ−φ) = 0, and the light intensity ratio I / Io is expressed by equation (6).

I / Io = 1/2 (6)
Θ when the amplitude of the fitting curve of the light intensity ratio I / Io is ½ is the main axis direction φ itself. In addition, there are two angles (φ−θ = 0, π / 2) in one cycle that I / Io = 1/2, but it is 90 degrees counterclockwise from the vibration surface of the fast wave. Since there is a slow vibration surface and there is a maximum amplitude between 90 degrees, before and after φ−θ = 0, the intensity ratio I / Io increases with respect to the rotation of the analyzer and is ½. Will be crossed.

  Therefore, if the relative phase difference δ is positive from the equation (3), the detected intensity ratio I / Io increases over 1/2 while before and after the main axis direction φ and the rotation angle θ of the analyzer coincide. Therefore, in the graph of FIG. 5, it can be seen that θ = π / 8 and coincides with φ, that is, the principal axis direction φ is 22.5 degrees (π / 8).

  However, when the relative phase difference δ is negative, conversely, θ is the main axis direction when I / Io decreases while straddling 1/2.

  Furthermore, a method for improving the measurement accuracy of the present invention will be described. When Expression (4) is solved for the relative phase difference δ, Expression (7) is obtained.

δ = arcsin (2 (I / Io) max−1) (7)
The graph of Formula (7) is shown in FIG. As is apparent from FIG. 12, as (I / Io) max approaches 0 or 1, the sensitivity of the change in the value of δ with respect to the error of (I / Io) max increases. In other words, it means that the measurement accuracy is lowered when the test object has a birefringence amount close to ± λ / 4. Therefore, for example, if a wave plate having a known relative phase difference, such as a λ / 4 plate, is inserted between the test object and the analyzer and measured, the relative accuracy that initially results in low measurement accuracy is obtained. Even a phase difference, that is, a value in the vicinity of δ = ± π / 2, is shifted to a relative phase difference with good measurement accuracy, and a measurement with high accuracy can be performed by subtracting a known relative phase difference from the result. In the present invention, this phase shift and the wavelength plate used to determine the direction of polarization rotation when expanding the measurement range are shared, and the range is expanded and the accuracy is improved simultaneously by one remeasurement. There is a feature that can be. When the wave plate is inserted and relative phase shift is performed, if the light transmitted through the test object is elliptically polarized light, the direction of the axis of the measurement point of the test object coincides with the direction of the wavelength plate axis. Since a relative phase difference cannot be generated, the direction of the axis of the wave plate needs to be inclined by 45 degrees with respect to the principal axis direction obtained in the first measurement.

  By the measurement method described above, the relative phase difference is determined in the range of −π / 2 ≦ δ ≦ π / 2, and the principal axis direction is determined in the range of 0 ≦ φ ≦ π. Here, a method for expanding the measurement range of the relative phase difference δ to −π ≦ δ ≦ π will be described.

  The measurement result shown in the graph of FIG. 5 where the relative phase difference is δ = + π / 4 actually has the same waveform as when δ = + 3π / 4. As is clear from equation (3), the maximum amplitude of I / Io is a function of sin δ and oscillates with a period of 2π, so the values of sin δ of δ = + π / 4 and δ = + 3π / 4 are equal. is there. The value of sin δ is shown in FIG. 6 in the range of −π ≦ δ ≦ π.

  However, the value of sin δ is the same between δ = π / 4 and δ = 3π / 4, but the rotation direction of the polarized light after passing through the test object is different. I can do it. FIG. 7 shows the value of sin δ in the range of −π ≦ δ ≦ π and the direction of polarization rotation. As shown in FIG. 7, when δ> 0 and clockwise, 0 ≦ δ ≦ π / 2, δ> 0 and counterclockwise, π / 2 ≦ δ ≦ π, and when δ <0, clockwise, −π / 2 ≦ When δ ≦ 0 and δ <0, the counterclockwise direction can be distinguished from −π ≦ δ ≦ −π / 2.

  Next, a method for obtaining the rotation direction of polarized light transmitted through the test object will be described. When a λ / 4 plate is inserted between the test object and the analyzer, the relative phase difference δ becomes δ + λ / 4. Therefore, this is set as δ ′, and the phase shifts as shown in FIG. At this time, the value of sin δ does not change when −π ≦ δ ≦ −π / 2 and 0 ≦ δ ≦ π / 2, but when −π / 2 ≦ δ ≦ 0 and π / 2 ≦ δ ≦ π, Therefore, 0 ≦ δ ≦ π / 2 and π / 2 ≦ δ ≦ π and −π ≦ δ ≦ −π / 2 and −π / 2 ≦ δ ≦ 0 can be distinguished from each other.

  Therefore, the detection range of the relative phase difference δ can be expanded from −π / 2 ≦ δ ≦ π / 2 to −π ≦ δ ≦ π.

  As described above, FIG. 13 shows a measurement technique in the birefringence measurement of the present invention and a flow of image processing for improving measurement accuracy.

  First, input the measurement conditions. The measurement conditions are the wavelength λ of the light source used for measurement, the number of image samplings associated with the rotation pitch of the analyzer, the measurement range, and the like.

  Next, a table for changing the luminance signal of the CCD to the light intensity is created. However, as long as the characteristics of the CCD do not change, it is not necessary to create each time once it is created.

  Next, the analyzer is rotated at a pitch according to the measurement conditions, and a blank image at each angle is accumulated in the PC.

  Next, the test object is arranged, and the test object image is accumulated in the PC under the same conditions as the blank image capturing.

  Next, fitting is performed by Equation (3) with respect to the change of the light intensity ratio I / Io at each rotation angle of the analyzer from the accumulated blank image and the test object image, and the transmittance error coefficient T is determined by the waveform. Based on the deviation amount from the light intensity ratio I / Io = 1/2 of the average Iave between the maximum value Imax and the minimum value Imin, the calculation is performed according to the equation (1). The light intensity ratio I / Io is corrected from the obtained transmittance error coefficient according to the equation (2).

  Next, the birefringence amount and the calculation condition in the principal axis direction are input. The calculation condition is the number of area divisions within the measurement range, and the maximum division number is the number of pixels of the CCD. When the number of divisions is smaller than the total number of images, calculation is performed based on an average value or a mode value in the divided area.

  Next, text data and visual image data are created and displayed for the measurement position, the amount of birefringence, and the principal axis direction.

  Here, when the amount of birefringence of the test object is in the vicinity of ± λ / 4 and measurement with high accuracy is required or when the measurement range needs to be expanded, there is a gap between the test object and the analyzer. A wave plate having a known relative phase difference such as λ / 4 is inserted, and the process is repeated from the capture of the test object image again. When this accuracy and range expansion process is performed, highly accurate data is adopted for each measurement point based on the previous measurement results and measurement results after re-measurement after the waveplate is inserted. Create and display data.

  Next, the measurement data is stored in the PC, and the measurement is completed.

  1 shows a first embodiment of a birefringence measuring method and a measuring apparatus using the same according to the present invention. As a test object, the amount of birefringence and the principal axis direction are naturally unknown, but in this embodiment, it is desirable that the amount of birefringence and the principal axis direction are accurately known for the purpose of verification of the present invention. An example in which a λ / 4 plate of quartz is used as a test object is shown below.

  The measurement light source is a halogen white light source, and a band-pass filter having a wavelength of 540 nm is attached to the light receiving camera. Therefore, the amount of birefringence of the test object is 135 nm, and the direction of the main axis is aligned to one side.

  In addition, when the irradiation light to the test object is slightly elliptical polarized light deviating from the circularly polarized light, the direction dependency of the test object arrangement occurs. Table 1 shows the measurement results at a measurement point having four birefringence amounts rotated by 45 degrees to 45 degrees.

  The luminance signal in Table 1 is obtained by calculating the birefringence amount using a luminance signal of 256 gradations from 0 to 255 with an 8-bit CCD camera. The light intensity is obtained by converting the luminance signal into light intensity according to the light intensity conversion table of FIG. 9 and calculating the birefringence amount. Further, the transmittance error coefficient is calculated and the result of correction is shown in the column of transmittance error correction. The value of the transmittance error coefficient T at this time is shown in Table 2. As is clear from the transmission error correction results in Table 1, the birefringence value calculated from the data of only the light intensity conversion varies greatly depending on the arrangement direction of the test object. The transmittance of the specimen is taken into account, and the error due to the deviation of the incident light from the circularly polarized light is also reduced, indicating a relatively close value in any direction. Furthermore, because ghost light easily enters the light receiving camera because of surface measurement, a field stop is placed between the test object and the light receiving camera, and unnecessary light is removed to show the results of improved measurement accuracy. 1 is shown in the ghost removal column. Further, the result of considering the extinction ratio of the measurement optical system in which the black level of the light receiving camera is increased by one gradation from the extinction ratio measured in advance is shown in the extinction ratio correction column of Table 1.

  FIG. 14 shows the amount of birefringence over the entire surface of the test object in this example, and FIG. 15 shows the principal axis direction. FIG. 14 is a three-dimensional bird's-eye view of the birefringence amount, and shows the birefringence amount in the height direction. As can be seen from the figure, the value of λ / 4 is shown over the entire surface of the test object. Further, it can be clearly seen that the main axis direction in FIG. 15 has the axes aligned in one direction because the test object is a wave plate. In FIG. 15, in order to make the information in the main axis direction for each pixel unit easy to see, the average value for each region is reduced and displayed.

  The 2nd Example of the birefringence measuring method of this invention and a measuring apparatus using the same is shown. FIG. 16 shows a measurement example of a test object having a relative phase difference distribution exceeding λ / 4. The horizontal axis of FIG. 16 indicates the measured pixel, and the vertical axis indicates the amount of birefringence, which is measured by the same measurement method as in Example 1 and shows a cross section thereof. In the figure, the x marks indicate the measurement results when the λ / 4 plate is not inserted to expand the measurement range and improve accuracy. It can be seen that as the birefringence amount approaches 135 nm, the sensitivity of the relative phase difference with respect to the luminance signal of the CCD increases, so that the measurement value becomes more discrete and the measurement accuracy deteriorates.

  Further, it can be seen that the birefringence amount is 135 nm, that is, δ = λ / 4 in the vicinity of the pixel number 50, and is the boundary data of the measurement range of −λ / 4 ≦ δ ≦ λ / 4. Therefore, it can be easily estimated that either the measurement data (a) of pixel numbers 0 to 40 or the measurement data (b) of pixel numbers 55 to 100 in the figure exceeds the measurement range.

  Next, by inserting a λ / 4 plate between the test object and the analyzer, and rotating simultaneously so that the optical axis of the λ / 4 plate and the transmission axis of the analyzer are maintained at an angle of +45 degrees. The obtained intensity ratio change is shown in FIG. 17 (a) and 17 (b) correspond to certain measurement points in the measurement region (a) and (b) in FIG. 16, and the black circles in the graph indicate the measurement without inserting the λ / 4 plate. The white square marks are measured values with the λ / 4 plate inserted. In FIG. 17A, the intensity ratio does not cross I / Io = 1/2 because the amplitude of the intensity ratio shifts the phase by λ / 4 with respect to the rotation of the analyzer, but in FIG. 17B, Since I / Io = 1/2 is crossed, the measurement point (b) exceeds the measurement range of λ / 4, and in fact, the distribution is turned back at 135 nm as indicated by the □ mark in FIG. I understand that

  Further, the amount of birefringence measured with the phase shifted by λ / 4 is indicated by a black dot. From the calculation reason shown in FIG. 12, it can be seen that the measurement accuracy is deteriorated in the region where the birefringence amount is close to 0, but conversely, it is improved in the region close to 135 nm.

The block diagram of the measuring apparatus using the birefringence measuring method of this invention. Explanatory drawing which shows the birefringence measuring method of this invention. Explanatory drawing which shows the shape of the test object which can be measured by this invention. Explanatory drawing which shows the picked-up image of this invention. Explanatory drawing which shows the change of the light intensity ratio of the blank image and test object image with respect to rotation of an analyzer. Explanatory drawing which shows the measurement range of a relative phase difference. Explanatory drawing which shows a relative phase difference and the rotation direction of polarized light. Explanatory drawing which shows the mode of the polarized light which shifted the relative phase difference. Explanatory drawing which shows the light intensity conversion linearity ((gamma)) of a CCD camera. Explanatory drawing which shows the transmittance | permeability error correction coefficient. Explanatory drawing which shows the light intensity ratio correct | amended by the transmittance | permeability error correction coefficient. Explanatory drawing which shows the precision of a measurement. Explanatory drawing which shows the process process of the birefringence measuring method of this invention. FIG. 2 is a bird's-eye view showing the amount of birefringence in Example 1. FIG. 3 is a zenith view showing the main axis direction of the first embodiment. FIG. 6 is a diagram showing the amount of birefringence in Example 2. The figure which shows the change of the intensity ratio in the state which shifted the phase of Example 2. FIG.

Explanation of symbols

δ Relative phase difference φ Main axis direction θ Angle of rotation of the analyzer Io Light intensity incident on the test object I Light intensity after transmitting the test object T Transmission error correction coefficient

Claims (20)

The light beam from the light source is made substantially uniform by the diffusion plate and is made circularly polarized light. A plurality of images recorded for each pixel by a CCD camera is used as a first image, and changes in brightness appearing by rotating the linearly polarizing plate with the same rotation width without inserting the test object for each pixel. The plurality of images recorded in the second image is used as a second image, and the first image and the second image are compared for each pixel, so that the amount of birefringence and the optical axis of the test object are measured at a plurality of locations in the measurement surface. In the desired birefringence measurement method,
Convert at least the first and second images from the luminance signal of the CCD camera into light intensity, and perform correction processing based on the transmittance of the test object, the degree of polarization of the light beam irradiating the test object, and the extinction ratio. A birefringence measuring method characterized by the above.   A plurality of images obtained by the first measurement of irradiating the object with circularly polarized light and recording the change of light and dark appearing for each pixel by the CCD camera by rotating the linear polarizing plate with respect to the light beam transmitted through the object. Is the first image, and in the measurement area where each pixel is the minimum unit, the birefringence amount of the test object and the direction of the optical axis are determined from the waveform of the change in the received light intensity with respect to the rotation angle of the linear polarizing plate at a plurality of positions In the birefringence measurement method determined in (1), a wave plate having a known phase difference is inserted between the test object and the linear polarizing plate, and the optical axis of the wave plate and the transmission axis of the linear polarizing plate are kept at a predetermined angle at the same time. A plurality of images obtained by the second measurement in which a change in brightness appearing by rotation is recorded for each pixel by a CCD camera is used as a second image, and the first image and the second image are compared for each pixel. Birefringence and optical properties of specimens Birefringence measuring method and measuring the direction.   The waveform of the intensity change of the transmitted light obtained in the first measurement is compared with the waveform obtained in the second measurement, and the measured value of the birefringence obtained in the first measurement is the actual value of the test object. When it is determined whether the birefringence amount exceeds the measurable birefringence range due to periodic changes to the birefringence amount, and exceeds the measurable birefringence range In addition to expanding the measurement range by adding an amount according to the periodic change to the measured birefringence amount, the birefringence value in the region where the measurement accuracy is low and the birefringence value obtained by the first measurement is low 3. The birefringence measurement method according to claim 2, wherein when the quantity is a quantity, a value obtained by subtracting the phase difference of the inserted wave plate from the second measurement result is used as a birefringence quantity at the measurement point.   4. The birefringence measuring method according to claim 2, wherein the wave plate inserted in the second measurement is a λ / 4 plate.   5. The birefringence measuring method according to claim 2, wherein the transmission axis of the linearly polarizing plate and the fast axis of the wave plate inserted in the second measurement are inclined by 45 degrees.   The light from the light source is made uniform by the diffusion plate, and the light and dark changes appearing by rotating the linear polarizing plate without inserting the test object after the light beam is made circularly polarized by using the phase difference plate, and the CCD camera A plurality of images obtained by the first measurement recorded every time are used as the first image, the test object is inserted between the circularly polarizing plate and the linearly polarizing plate, and the linearly polarizing plate is the same for the light beam transmitted through the test object. A plurality of images obtained by the second measurement in which the change in brightness appearing by rotating with the rotation width is recorded for each pixel by the CCD camera is defined as the second image, and the first image and the second image are defined for each pixel. In the birefringence measurement method for determining the amount of birefringence of the test object and the direction of the optical axis at a plurality of locations in the measurement surface by comparing, the first and second images are converted from the luminance signal of the CCD camera to the light intensity The transmittance of the specimen and the specimen Correction processing is performed based on the degree of polarization and extinction ratio of the light beam to be inserted, a wave plate having a known phase difference is inserted between the test object and the linear polarizing plate, and the optical axis of the wave plate and the transmission axis of the linear polarizing plate are set. A plurality of images obtained by a third measurement in which changes in brightness appearing by rotating simultaneously while maintaining a predetermined angle are recorded for each pixel by a CCD camera is defined as a third image, and the first image is defined for each pixel. And measuring the birefringence amount and the direction of the optical axis of the test object by comparing the second image with the third image.   The waveform of the intensity change of the transmitted light obtained by the second measurement is compared with the waveform obtained by the third measurement, and the measured value of the birefringence obtained by the second measurement is the actual value of the test object. When it is determined whether the birefringence amount exceeds the measurable birefringence range due to periodic changes to the birefringence amount, and exceeds the measurable birefringence range Includes an amount corresponding to a periodic change in the amount of birefringence to be measured to expand the measurement range, and the birefringence value obtained by the second measurement is a region where the measurement accuracy is low. 7. The birefringence measurement method according to claim 6, wherein when the quantity is a quantity, a value obtained by subtracting the phase difference of the inserted wave plate from the third measurement result is used as a birefringence quantity at the measurement point.   The birefringence measuring method according to claim 6 or 7, wherein the wave plate inserted in the third measurement is a λ / 4 plate.   9. The birefringence measuring method according to claim 6, wherein the transmission axis of the linearly polarizing plate and the fast axis of the wave plate inserted in the third measurement are inclined by 45 degrees.   The dark level according to the extinction ratio of the measurement optical system and the dark level of the CCD camera are incorporated in a conversion table or conversion function for converting the luminance signal of the CCD camera into light intensity. 10. The birefringence measuring method according to any one of 1 to 9.   The first and second images are converted from the luminance signal of the CCD camera into light intensity, corrected based on the transmittance of the test object, the degree of polarization of the light beam that irradiates the test object, and the extinction ratio. A wave plate with a known phase difference is inserted between the specimen and the linear polarizing plate, and changes in brightness appear by rotating simultaneously while maintaining the optical axis of the wave plate and the transmission axis of the linear polarizing plate at a predetermined angle. When a plurality of images obtained by the third measurement recorded for each pixel by the CCD camera are used as the third image, the first image and the second image, or the first image and the third image 11. The birefringence measurement method according to claim 1, wherein the comparison is performed for each corresponding pixel so as not to be affected by illumination unevenness and noise depending on a measurement position by a CCD or the like.   The birefringence measuring method according to claim 1, wherein the light source uses a primary color filter for white or white.   The birefringence measurement method according to claim 1, wherein the light source in the previous period has a bright line, and the CCD camera can receive light in color.   The birefringence measurement method according to claim 1, wherein the measurement by the measurement method is controlled from a PC monitor screen.   The first and second images are converted from the luminance signal of the CCD camera into light intensity, corrected based on the transmittance of the test object, the degree of polarization of the light beam that irradiates the test object, and the extinction ratio. A wave plate with a known phase difference is inserted between the specimen and the linear polarizing plate, and changes in brightness appear by rotating simultaneously while maintaining the optical axis of the wave plate and the transmission axis of the linear polarizing plate at a predetermined angle. When a plurality of images obtained by the third measurement recorded for each pixel by the CCD camera are used as the third image, the rotation of the linear polarizing plate at the measurement point is used to correct the transmittance and the polarization degree of the test object. Correction is performed based on the deviation of the average value of the maximum value and the minimum value of the change in the light intensity ratio between the first image and the second image or the first image and the third image from the desired value. The birefringence measuring method according to claim 1, wherein The first and second images are converted from the luminance signal of the CCD camera into light intensity, corrected based on the transmittance of the test object, the degree of polarization of the light beam that irradiates the test object, and the extinction ratio. A wave plate with a known phase difference is inserted between the specimen and the linear polarizing plate, and changes in brightness appear by rotating simultaneously while maintaining the optical axis of the wave plate and the transmission axis of the linear polarizing plate at a predetermined angle. When a plurality of images obtained by the third measurement recorded for each pixel by the CCD camera is used as the third image, the correction of the transmittance and the polarization degree of the test object is performed with respect to the rotation of the analyzer at the measurement point. When the average value of the maximum and minimum values of the change I / Io in the ratio I / Io between twice the light intensity Io of the first image and the light intensity I of the second or third image is corrected by the following equation: The amount of birefringence and the direction of the optical axis can be obtained from the obtained light intensity ratio (I / Io) ′. Birefringence measuring method according to any one of claims 1 to 15, characterized in.
T = Iave / 0.5
(I / Io) ′ = [{I / Io + (1−T) / 2} −0.5] /T+0.5   In addition to the measurement optical path, it can be converted into a fiber having a diffusing plate and / or a polarizing plate at the tip branched from the measurement optical path, and the optical axis passing through the exit surface of the fiber coincides with the optical axis of the CCD camera. The birefringence measurement method according to claim 1, wherein the birefringence measurement method is not performed.   A birefringence measuring apparatus using the birefringence measuring method according to claim 1.   An optical element characterized in that the birefringence amount and the direction of the optical axis are measured by using the birefringence measuring device according to claim 18, and the optical performance is compensated by a processing condition or a correction element.   An optical system using the optical element according to claim 19.

Images